Here is one answer:
The amount of gain reduction is determined by ratio: a ratio of 4:1 means that if input level is 4 dB over the threshold, the output signal level is 1 dB over the threshold. The gain (level) has been reduced by 3 dB: Threshold = −10 dB.
Comments
It's very easy to make this subject confusing to those who are n
It's very easy to make this subject confusing to those who are not well versed in the use of compressors. It's not helped by the almost universal use of displaying an output/input gain curve where unity gain (no compression) is a 45degree angle straight line.
When I've been involved in teaching this topic, I show the conventional diagram and then show it redrawn as a gain reduction diagram, where unity gain is a horizontal line at 0dB which then rolls off to a slope at a certain threshold, resembling a low pass filter. Eyes light up when I show this, as it is clearly an easier thing for newcomers to the topic to take in.
Boswell, post: 446147, member: 29034 wrote: It's very easy to ma
Boswell, post: 446147, member: 29034 wrote: It's very easy to make this subject confusing to those who are not well versed in the use of compressors. It's not helped by the almost universal use of displaying an output/input gain curve where unity gain (no compression) is a 45degree angle straight line.
When I've been involved in teaching this topic, I show the conventional diagram and then show it redrawn as a gain reduction diagram, where unity gain is a horizontal line at 0dB which then rolls off to a slope at a certain threshold, resembling a low pass filter. Eyes light up when I show this, as it is clearly an easier thing for newcomers to the topic to take in.
My eyes are lit if you care to explain... although now that you said the LPF thing...I think I understand what you mean. And the knee, is just the slope of the rolloff, until it connects back to unity gain. The knee could be a small curve...or it could be a hard pivot...yes?
I still find compressors to be confusing. I read about them. I read manuals for hw compressors, I read the manuals for my plug-in compressors....and I still don't quite get it. There is a piece of knowledge that I'm just not getting.
So the ratio of 4:1 in your example audiokid , would y0u use that ratio and leave a +1 over threshold? Or was it just random numbers? And when you are using the comp, are you basing your settings on the db's from your meters? In other words, are you choosing the 4:1 bc of how it sounds? Or based on the fact that you know this track is 4db over threshold?
"Make up" takes stuff that is too low in volume and boosts it?
I think sometimes I see another knob with 3 numbers, e.g. 5:1:1, and I don't know what those numbers are indicating. I manage....but I honestly don't understand what they mean.
And some people don't just compressors for gain reduction. I think I'm fairly close to understanding how to use them well for that purpose. But some people use them as an artistic tool...and I have no idea what they are doing, to accomplish that.
Parallel Compression, oddly enough, I have always understood that it's simply mixing compressed and uncompressed, or differently compressed. Almost like mixing a wet and dry track. It's the simpler compression part that still mystifies me a little.
I guess a large question I have is are you guys setting them by the numbers, or by ear? Maybe if I understood how to set it by the numbers, the ear part would just come to me.
Brother Junk, post: 446183, member: 49944 wrote: I guess a large
Brother Junk, post: 446183, member: 49944 wrote: I guess a large question I have is are you guys setting them by the numbers, or by ear? Maybe if I understood how to set it by the numbers, the ear part would just come to me.
No, never set by numbers alone, unless the compressor is being used as a limiter (infinite ratio compression).
As with any tool like this, you have to let the artistic angle dictate the engineering. You as the producer have to make an assessment and tell yourself as an engineer that compression is required, how much (ratio), at what level it should become operative (threshold), and then a guess at how the attack and release times should be set. The engineer tries these settings and the producer adjusts them as required. Don't be afraid to throw out the compressor if you can't decide whether it makes an improvement - I've have to deal with too many tracks that have been brought in for mixing with inappropriate compression added during recording.
It also makes a big difference whether you are talking about compression of a single track such as a vocal prior to feeding it into a mix, or whether you are dealing with compression/limiting of the two-track mixed output.
Boswell, post: 446184, member: 29034 wrote: No, never set by num
Boswell, post: 446184, member: 29034 wrote: No, never set by numbers alone, unless the compressor is being used as a limiter (infinite ratio compression).
Got it. However, the fact that you are saying not to use numbers alone, implies that you are using numbers to help you get to some sort of starting point? So, are you using numbers calculated from the meters to get you to a baseline, and then you play artistically? Or how involved are the actual numbers, that you said not to "set by numbers alone?"
Were my other assessments of how it works at least close to correct?
Boswell, post: 446184, member: 29034 wrote: and then a guess at how the attack and release times should be set.
So is that where the artistic angle comes in? Because it will naturally adjust the groove?
Boswell, post: 446184, member: 29034 wrote: I've have to deal with too many tracks that have been brought in for mixing with inappropriate compression added during recording.
This may sound silly, but I believe part of my confusion comes from not seeing them used much, bc the studio I've mostly gone to...they really know their stuff. They are elite. I'm amazed that I've been in the same room with them. And I've never seen them use a compressor. They have hw, but he says he often doesn't use it. Sometimes he'll throw a stock pi in, but, but it's just a quick default or saved setting. He does the "final" mix later on...so I've never gotten to see someone who really knows how to use a compressor, use a compressor.
And I feel like a lot of the people on youtube who "truly" know how to use one...they are often too advanced to meet my understanding. And the other ones, sometimes they understand it less than I do, yet they are offering on line lessons....
I don't find I need them that much, and when I do, I feel like I'm "ok" with it, but I don't feel like I truly get it. It's not just bam, bam, bam for me.
I guess that's why I'm wondering how you guys may use the numbers to get to a baseline?
Brother Junk, post: 446183, member: 49944 wrote: So the ratio of
Brother Junk, post: 446183, member: 49944 wrote: So the ratio of 4:1 in your example audiokid , would y0u use that ratio and leave a +1 over threshold? Or was it just random numbers? And when you are using the comp, are you basing your settings on the db's from your meters? In other words, are you choosing the 4:1 bc of how it sounds? Or based on the fact that you know this track is 4db over threshold?
Boswell, post: 446184, member: 29034 wrote: No, never set by numbers alone, unless the compressor is being used as a limiter (infinite ratio compression).
As with any tool like this, you have to let the artistic angle dictate the engineering. You as the producer have to make an assessment and tell yourself as an engineer that compression is required, how much (ratio), at what level it should become operative (threshold), and then a guess at how the attack and release times should be set. The engineer tries these settings and the producer adjusts them as required. Don't be afraid to throw out the compressor if you can't decide whether it makes an improvement - I've have to deal with too many tracks that have been brought in for mixing with inappropriate compression added during recording.
It also makes a big difference whether you are talking about compression of a single track such as a vocal prior to feeding it into a mix, or whether you are dealing with compression/limiting of the two-track mixed output.
Bos said it well. (y)
Compression and the types of compressors used has a wide scope of uses.
Analog compressors on vocals:
LA2A optical type compressors work well tracking vocals. I use UA LA2A's to gently level a vocal. I have done tests (ITB vs OTB) and these particular compressors always sound noticeably better tracking vs a LA2A plugin emulation, including using the real deal in a mixing hybrid approach as well.
There is something extra special when used during tracking that I have never been able to reproduce after a track is ITB or hybrid/ round trip.
That being said: The UA LA2A has a nice big VU meter on it. Although I look at it, and use it for a guide line, I generally use my ears for the desirable effect I am pursuing. I come from the old days where VU meters don't always represent certain results quickly. Meters on those compressors are more of an indicator that something is working but it may have happened already. So I look at the meter but close my eyes too.
Regardless of analog or digital compressors, once I find the effect I'm looking for, I stay close or back off at the point of obvious notice. When it comes to effect processing (compression), I generally follow a rule that once you hear something changing, baby steps come next.
audiokid, post: 446186, member: 1 wrote: Bos said it well. (y) H
audiokid, post: 446186, member: 1 wrote: Bos said it well. (y)
He always does. I wish I could have 24 hours with that brain of his....and just copy/paste all the understanding lol...
**Edit** Maybe the above needs context, since I'm relatively new here. I do a lot of back reading. I've gone through so much info and idle chatter between you long-timers that you probably forgot even existed. 'Boz is always well spoken.
No context needed on the brain part. I'd like to copy and paste all of his knowledge onto my brain. My biggest "career" background has been in business. Making $$. Just take this 'thing' and make more money with it.
If someone were able to figure out how to do something like that....take Boz's career brain info, and give it to someone else. Everyone who has something to offer, and the person who figures out how to do it would make a fortune. I 'get that it would "sterilize" the creative world....but, it's just what I've been programmed to see...how to make $$$.
And it would make so many people rich. Omg, if you were able to license (in other words, make it theft proof), people like you and Boswell, could make $50k a crack, probably much more (I could get you more ;)
If any of you figures that out, I'd like a cut lol. I've been reading way too much sci-fi.
Like Chris, I like the LA2A especially well on vocals going in a
Like Chris, I like the LA2A especially well on vocals going in and just as a 'leveler'. When you are using mics that amplify the slightest nuance in tone, volume and the space around the mic, it's always good to have a handle on the signal going down. NOT LIMITING.....Then once its in, at the mix you really have a lot of choices.....like slamming it particularly hard in the digital world...........This is when I really like the UAD LA2A with the gain and the peak reduction way way up there giving that nice even grit to a voice......just like the old days with no noise from the device!
Davedog, post: 446309, member: 4495 wrote: This is when I really
Davedog, post: 446309, member: 4495 wrote: This is when I really like the UAD LA2A with the gain and the peak reduction way way up there giving that nice even grit to a voice......just like the old days with no noise from the device!
Cool, Dave. I'll have to try your UA /UAD la2a combo approach, with more love.
This implies the original signal level is -6db which is 4db over
This implies the original signal level is -6db which is 4db over the -10db threshold. ;)